---

Paul Ferris — Windows 2000: Pizza Express

[ The opinions expressed by authors on Linux Today are their
own. They speak only for themselves and not for Linux Today.
]

By Paul Ferris, Staff
Writer

Back in the early 80’s when the IBM PC was just taking off, I
was working as a delivery guy for Dominos pizza on my college
campus. One of my managers from time to time wore what looked like
an old Dominos shirt that had a badge that instead said “Pizza
Express”.

It was a while before I noticed the difference between that and
a stock uniform because the color and placement of the company logo
on the shirt were close to identical.

He explained to me that at one time there had been a lawsuit
between Dominos and the folks that made Domino sugar, but that it
had since been resolved. During that lawsuit the shirts were
changed and all new stores bore the name Pizza Express instead of
Dominos.

It was what he said next that I found particularly interesting.
After that Dominos experimented with something really crazy using
the Pizza Express name brand. They opened stores in competition
with themselves in several places, just to see what the effect
was.

The idea was simple: become your own competition, before someone
else did it for you.

GM does something similar by pitting Pontiac, for example,
against Chevrolet. By competing with themselves, they enter a
marketplace in two or even more areas instead of one. The overhead
is increased a bit, but it’s not like the overhead that would be
present in two competing companies.

I think that it’s for similar reasons that Microsoft is motivated to change
the Win32 API (Application Programming Interface). For Windows 2000
they are telling developers to re-write everything to run on the
new operating system.

Most of this, we are told, is to take advantage of the new, but
very late, Active Directory services. Some of it, I’m sure, has to
do with the fact that somewhere in Redmond the clue meter
registered above zero for a minute, and somebody decided that with
Windows 2000 (professional) programs cannot just replace whatever
system DLL with whatever felt good at installation time.

A lot of people are saying that the timing for all of this
rewriting is crazy. The strain is coming at the same time as Y2K
code fix issues. Now Microsoft wants developers to devote valuable
resources to W2K rewrites.

But playing Bill Gates advocate here (loosely): It makes sense,
in a twisted sort of way, because these developers otherwise might
devote that time to a totally different API: Unix/Linux.

Remember that developer share is as important to Microsoft as
market share. I can’t imagine, what with Java and Linux taking a
larger and larger slice of that pie, that they don’t wish that
there was some way to halt or reverse the trend. By making
developers choose between Microsoft and Linux for this new software
programming time, they are competing with themselves.

Cool is rumored to be a Java migration tool, to give developers
a way to migrate back to C++ and therefore back into the Microsoft
developer camp. It would make sense, since Microsoft is afraid of
the Java development community in the worst way.

There are other competitive issues as well. There are rumors
that somebody is preparing a Windows NT clone. Microsoft in the
past has fretted that somebody might reverse engineer the Win32 API
and create a Windows competitor. The Wine project will eventually provide a
developer migration bridge for the Win32 API anyway. The present
one at least. There is at least one company that I can think of
that has a Unix migration product (currently without any Linux
support) that could make things extremely interesting if they
wanted to.

Microsoft has at least these factors to motivate it to change
the API and keep the developer ball in their court.

At least, that’s how they might see it to their advantage. One
of the things that I find appalling is the marketing phrase “Oh,
that’s old technology”. It’s a spin tactic that is
backfiring from what I can tell.

Old technology may be boring, but it also tends to have fewer
bugs in it than New Technology, if you get my drift. Open Source
and Free Software old technology, from what I see,
although boring from a marketing standpoint, have brought to the
table many benefits that yawn, excuse me, sell it very
nicely. There’s that stability (yawn) and efficiency (zzz) factor
to consider. These factors don’t usually rear their ugly head until
you mention the words “mission critical”, but lately they’ve been
getting Linux a lot of good press.

Oh, and let’s not forget that the Unix API was supposed to die
when Windows took over. Fat chance now. It’s suddenly achieved a
life that a few short years ago no one would have believed could
happen.

Microsoft may be demonstrating just how dangerous “new
technology” can be to future marketing plans with their previous
track record. Now they want to implement another slice of “new
technology”: Windows 2000. Hey, it has a brand
new!
API as another selling point. Don’t worry about us
cleaning up the last several APIs that have blown in and out of
Microsoft town, this new one will be even better!

I can’t imagine that every developer is going to jump on the W2K
bandwagon just yet. Microsoft keeps dragging their feet on the
release date, and they have shown an amazing ability to slow their
development times down lately.

It leaves a gap for a competitor to arrive at any time. If
something were to happen to Wine for example, where without extra
binaries people could easily run windows programs on their Linux
systems today, then it would change the whole problem virtually
overnight.

Then, there’s the problem of Free Software development.
Proprietary software vendors that stall Linux migration today can
expect the vacuum to be filled by an Open Source or Free Software
alternative. By demanding that commercial developers devote their
resources only to W2K development, Microsoft may actually be giving
the Free Software movement more time to complete their software
portfolio.

The Windows API is changing again, and it could be that a lot of
companies have no choice but to support this trend. Very likely, a
lot of proprietary software developers see this as inevitable. They
must do the bidding of their only source of tools: Microsoft. Some
may see it as a way to increase revenue. New system software
requires that customers go out and upgrade just so that the
existing programs will run.

But I see it a different way: Microsoft is trying to compete
with Microsoft. Given what a slow target they have become, this may
be some very bad timing.

At least that’s my take, I’d be interested in yours.